On International Migrants Day, the Court of Justice ruled on two appeals in action for damages brought against Frontex: the case WS and Others v European Border and Coast Guard Agency (Frontex), and the case Alaa Hamoudi v European Border and Coast Guard Agency (Frontex). It upheld both appeals, set aside the judgments of the General Court (in its entirety for Hamoudi and largely for WS) and the cases were reverted back to the General Court. This means that we will (yet again) have to wait for new rulings. This blogpost will analyse the appeal in the Hamoudi case.

Background to the case: an unattainable burden of proof

The Hamoudi case was decided in December 2023 and concerned a Syrian national who entered Greece to seek asylum. On the same day, the Greek police intercepted him and others, seized their phone, and returned them to sea. The applicant claims that the group was overflown on two occasions by a Frontex-operated surveillance aircraft. Consequently, he sought compensation from Frontex under Article 268 TFEU for damages suffered during and after the pushback operation, arguing that the Agency failed to comply with obligations arising from Frontex Regulation (in this case Regulation (EU) 2019/1896) and the Charter of Fundamental Rights.

The General Court, disappointingly yet not surprisingly, dismissed the action, arguing that the applicant was ‘not manifestly capable of demonstrating, to the requisite standard, the events relating to the alleged incident of 28 and 29 April 2020 which led to the alleged damage’ (para. 57), nor its presence in those incidents. Therefore, without looking at the other conditions for extra-contractual liability, it concluded that the actual damage has not been demonstrated.

The decision of the General Court has been criticised for failing to provide (once more) clarification on the cumulative conditions that must be met in an action for damages (see here), and for requesting an extremely high and unrealistic burden of proof (see here and here). In fact, the applicant submitted a written declaration, a Bellingcat article on the event and several photographs taken from YouTube videos documenting the pushback However, these elements were deemed insufficient to demonstrate the applicant’s involvement in the alleged incident. Thus, academics argued (see here and here) that the Court should have followed the European Court of Human Rights (ECtHR) approach that requires that either the burden of proof be shared between the parties involved in the case (as has been done in Alkhatib and Others v Greece) or reversed, when certain facts are within the exclusive knowledge of the defendant.

The applicant appealed the decision, and Advocate General (AG) Norkus delivered his opinion on 10 April 2025. In his opinion, the AG noted that according to the Court of Justice and the ECtHR case-law, a reversal of the burden of proof can take place if a number of cumulative conditions are fulfilled: (1) the claimant must successfully adduce prima facie evidence to support their case; (2) there must be a clear or structural imbalance in accessing the evidence; (3) not reversing the burden of proof would undermine the claimant’s rights under EU law. The AG then proceeded in noting that there is a presumption of the claimant being at ‘a disadvantage in adducing evidence’ in EU anti-discrimination legislation and collective expulsion (para. 61). He, however, noted that this presumption cannot apply to actors ‘other than the authorities of a Member State’ (e.g. Frontex) due to the fact that these actors have limited powers in comparison to national authorities (para. 62). He claimed that for these actors (here Frontex), it is less evident and can thus not be presumed that they contributed to the difficulties for the applicant to obtain the evidence. As argued by Pirrello here, however, such an approach fails to reflect the concrete powers that the Agency has in the field. The AG ultimately suggested that the Court either rejects the appeal if it could be determined that Hamoudi did not provide prima facie evidence of damage, or annuls the order of the General Court and rules on whether the three conditions described above to reverse the burden of proof are applicable.

Decision of the Court of Justice: a glimpse of (limited?) hope

In the Hamoudi judgment (C-136/24 P), the Court of Justice set aside the decision of the General Court and stated that the General Cout had infringed the applicant’s right to effective judicial protection (Article 47 of the Charter of Fundamental Rights), by not adapting the burden of proof in light of the circumstances of the case. It added that it cannot be asked from victims of a pushback to demonstrate through conclusive proof that the pushback occurred and that they were present. Instead, it stated, that the evidence brought by the applicant (e.g. the press article and the witness statement) were sufficiently detailed, specific and consistent to constitute prima facie evidence, and thus to be subject to an adaptation of the burden of proof. It then noted that where prima facie evidence is produced (as it was in this instance), the General Court is required to investigate the case and obtain from Frontex all relevant information, to establish whether the pushback took place, and if the applicant was present.

Two main aspects are worth analysing further in detail. First, the conditions to reverse the burden of proof, and second, the assessment of prima facie evidence.

1. The reversal of the burden of proof: a simplified process to promote greater transparency of Frontex

The Court of Justice in its reasoning on the reversal of the burden of proof did not refer to the three conditions discussed in the opinion of the AG, namely (1) prima facie evidence; (2) clear or structural imbalance in accessing the evidence; (3) a risk of undermining the claimant’s rights under EU law if the burden of proof is not reversed. Instead, it essentially focuses on the first condition, while still admitting that not adapting the burden of proof would undermine Article 47 of the Charter (para. 106).

This simplification of the conditions to be met to adapt the burden of proof, would seem to allows parties to focus on establishing the first condition (prima facie evidence), without needing to bring arguments forward to show a clear or structural imbalance, and a risk of undermining the applicant’s rights under EU law. This would also mean that there is no need for the Court to clarify and assess the other conditions in similar cases (e.g. what is the threshold needed to assess an asymmetry?). However, the Court of Justice does not engage with the argument whether the presumption of reversal of burden of proof applicable in collective expulsion cases, can apply to an actor such as Frontex, leaving the question unanswered.

By only requiring prima facie evidence, in a similar manner as the presumption in EU anti-discrimination legislation and collective expulsion and refoulement cases (see the Opinion of AG para. 61), the Court of Justice lowers the burden of proof for applicants. As a result the General Court will likely be more often required to take further steps in investigating the case and requesting information from Frontex. This conclusion reached by the Court is highly significant and might be an initial response to tackle the lack of transparency of the Agency’s operations (see here and here). It is now up to the General Court to take a decision, and (hopefully) request further information from Frontex regarding the incident at stake.

2. Constituting prima facie evidence: witness statement or corroborating evidence?

When it comes to prima facie evidence, the Court stated that the evidence the applicant brought, was sufficiently detailed, specific and consistent to constitute prima facie evidence. It noted notably that the General Court erred in law in dismissing the witness statement as having little probative value (para. 117) and erred in law in dismissing the Bellingcat article which supported the prima facie evidence provided by the witness statement of Alaa Hamoudi (para. 125). The Court of Justice ultimately took a stronger stance than the AG, since the AG only said that the General Court might have set the bar too high (para. 65 of his opinion) without concluding on whether or not the applicant submitted prima facie evidence.

This is a positive development for applicants that wish to bring cases against Frontex before the Court of Justice. Two points remain however unclear. Firstly, how the General Court will ultimately assess whether a witness statement is sufficiently ‘detailed, specific and consistent’ (para. 123), and secondly, what constitutes prima facie evidence for the Court of Justice.

The press release of the Court of Justice on the case seems to imply that the witness statement and the press article were sufficiently detailed, specific and consistent to constitute prima facie evidence. The Court in its judgement also refers to ‘a body of consistent evidence’ (paras. 128 and 152), which seems to imply that the mere witness statement did not suffice but must be supported by other evidence. If that was the case, this might significantly limit the scope of the judgment, as it is extremely hard for individuals to obtain evidence of pushbacks (in the same form of the Bellingcat article). This would also ultimately put a high burden on external actors, such as civil society organisations and investigative journalists, to play an active role in monitoring the external borders of the EU, and report on incidents. This is highly problematic in the current situation of criminalisation of solidarity in the EU (see for example the report published by PICUM).

In other parts of the judgment, the Court of Justice seems to imply that the sole witness statement should be enough to constitute prima facie evidence. On several occasions, it notes that the General Court should have considered the witness statement as prima facie evidence (para. 119 and 122), and that the Bellingcat article merely supports the prima facie evidence provided by the witness statement (para. 125). This is a positive evolution of the case law, since it will be easier for individuals to bring cases, if their witness statement suffices to adapt the burden of proof.

Additionally, the Court of Justice notes that this interpretation would be in line with the case law of the ECtHR on collective expulsion, on the basis of which it is sufficient for the applicant to ‘provide a detailed, specific and consistent account of the events in issue’ (para. 109). However, when looking at the case-law of the ECtHR referred to by the Court of Justice, it seems that very often the ECtHR takes into account corroborating evidence to establish prima facie evidence (for more details see here). In N.D. and N.T. v Spain, for example, next to the coherent witness statement, the Court also took into account video-footage from journalists and other witness (paras. 27 and 86). Similarly, in M.H. and Others v Croatia, the ECtHR took into account a number of reports from civil society organisations, national human rights structures and international organisations regarding the situation at the Croatian border with Serbia and Bosnia Herzegovina (para. 270). Finally, in A.R.E. v Greece, the Court also examined national and international institutions’ reports on ‘pushbacks’ from the Evros region to Türkiye (para. 230). It is thus, unclear, whether the General Court will request supporting evidence in subsequent cases, or whether a sole witness statement will be deemed sufficient to constitute prima facie evidence.

Concluding thoughts: will Frontex finally be held liable?  

The burning question remains: will Frontex finally be held liable for violations of fundamental rights? The Court of Justice in its judgment does not provide an answer to this question, and – as had been criticised in respect of the judgment of the General Court – failed to engage with issues of joint liability. The Court does, however, offer some hope in a so-far gloomy landscape. In fact, it gives a strong statement that the action for non-contractual liability is possibly the only legal remedy capable of offering judicial protection against the actions of Frontex (para. 103). However, even that action has not proven to be successful so far to hold Frontex accountable (e.g., the WS and Others v Frontexcase from the General Court), due to the stringent conditions for extracontractual liability: unlawful conduct, sufficiently serious breach and causal link. It is highly probable that the outcome from the General Court in the Hamoudi case will not be satisfactory. In fact, even with an investigation from the General Court resulting in additional evidence received from Frontex on the incident, it will be extremely hard to meet the causality requirement between the alleged conduct of Frontex and the pushback. Therefore, while the judgment of the Court of Justice is a positive one, it might not conclude in a full victory of the applicant before the General Court.

Posted by Sarah Tas (University of Maastricht)

Sarah Tas is Assistant Professor in the Public Law Department, at Maastricht University. She works at the intersection of Administrative law, European law, fundamental rights protection and more specifically data protection. Sarah has extensively published on the supervision and accountability of Frontex and Europol, as well as their datafication. Her work also has societal significance; she has notably co-authored a targeted substitute impact assessment for the European Parliamentary Research Service, on a proposal strengthening Europol’s role in the fight against migrant smuggling and trafficking in human beings, and is also a member of Europol’s Consultative Forum